Temperature in SLMs: Impact on Incident Categorization in On-Premises Environments
Resumo
SOCs and CSIRTs face increasing pressure to automate incident categorization, yet the use of cloud-based LLMs introduces costs, latency, and confidentiality risks. We investigate whether locally executed SLMs can meet this challenge. We evaluated 21 models ranging from 1B to 20B parameters, varying the temperature hyperparameter and measuring execution time and precision across two distinct architectures. The results indicate that temperature has little influence on performance, whereas the number of parameters and GPU capacity are decisive factors.
Referências
Ibrishimova, M. D. (2019). Cyber incident classification: Issues and challenges. In Advances on P2P, Parallel, Grid, Cloud and Internet Computing.
Li, L., Sleem, L., Gentile, N., Nichil, G., and State, R. (2025). Exploring the impact of temperature on large language models:hot or cold?
Ogundairo, O. and Broklyn, P. (2024). Natural language processing for cybersecurity incident analysis. Journal of Cyber Security.
Renze, M. (2024). The effect of sampling temperature on problem solving in large language models. In Findings of the Association for Computational Linguistics: EMNLP 2024, page 7346–7356. Association for Computational Linguistics.
Severo, A., Lautert, D., Almeida, G., Kreutz, D., Rodrigo, G., Jr, L. P., and Bertholdo, L. (2025a). LLMs e engenharia de prompt para classificação automatizada de incidentes em SOCs. In Anais Estendidos do XXV SBSeg. SBC.
Severo, A., Lautert, D., Kreutz, D., Bertholdo, L., Pohlmann, M., and Quincozes, S. (2025b). Categorização de incidentes de segurança utilizando engenharia de prompts em LLMs. In Anais do XXV SBSeg, pages 256–272. SBC.
Wang, P.-H., Hsieh, S.-I., Chang, S.-C., Chen, Y.-T., Pan, J.-Y., Wei, W., and Juan, D.-C. (2020). Contextual temperature for language modeling.
Zhao, W. X., Zhou, K., Li, J., Tang, T., Wang, X., Hou, Y., Min, Y., Zhang, B., Zhang, J., Dong, Z., et al. (2023). A survey of large language models. arXiv preprint arXiv:2303.18223, 1(2).
Li, L., Sleem, L., Gentile, N., Nichil, G., and State, R. (2025). Exploring the impact of temperature on large language models:hot or cold?
Ogundairo, O. and Broklyn, P. (2024). Natural language processing for cybersecurity incident analysis. Journal of Cyber Security.
Renze, M. (2024). The effect of sampling temperature on problem solving in large language models. In Findings of the Association for Computational Linguistics: EMNLP 2024, page 7346–7356. Association for Computational Linguistics.
Severo, A., Lautert, D., Almeida, G., Kreutz, D., Rodrigo, G., Jr, L. P., and Bertholdo, L. (2025a). LLMs e engenharia de prompt para classificação automatizada de incidentes em SOCs. In Anais Estendidos do XXV SBSeg. SBC.
Severo, A., Lautert, D., Kreutz, D., Bertholdo, L., Pohlmann, M., and Quincozes, S. (2025b). Categorização de incidentes de segurança utilizando engenharia de prompts em LLMs. In Anais do XXV SBSeg, pages 256–272. SBC.
Wang, P.-H., Hsieh, S.-I., Chang, S.-C., Chen, Y.-T., Pan, J.-Y., Wei, W., and Juan, D.-C. (2020). Contextual temperature for language modeling.
Zhao, W. X., Zhou, K., Li, J., Tang, T., Wang, X., Hou, Y., Min, Y., Zhang, B., Zhang, J., Dong, Z., et al. (2023). A survey of large language models. arXiv preprint arXiv:2303.18223, 1(2).
Publicado
08/12/2025
Como Citar
POHLMANN, Marcio; SEVERO, Alex; ALMEIDA, Gefte; KREUTZ, Diego; HEINRICH, Tiago; PEREIRA, Lourenço.
Temperature in SLMs: Impact on Incident Categorization in On-Premises Environments. In: ESCOLA REGIONAL DE REDES DE COMPUTADORES (ERRC), 22. , 2025, Porto Alegre/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 151-157.
DOI: https://doi.org/10.5753/errc.2025.17813.
